Satellite Data for the Oceans Joe Sienkiewicz, NOAA Ocean Prediction Center College Park, Maryland CEOS-SIT Thank you for the opportunity to address this important committee. I want to represent two perspectives in this brief talk. The first is of an operational marine meteorologist with wind warning responsibility over large ocean areas with complex weather and interactions. The second is as a member of a training team that shares knowledge about remote sensing capabilities for oceans for key parameters of winds, waves, and imagery.

anticipated Latency – as low as possible Data user perspective Data formats accepted anticipated Latency – as low as possible Thanks to many of you in this room and the governments and agencies around the world we live in an age of remarkable remote sensing capabilities from earth observing satellites in geostationary and low earth orbits. This geocolor image from GOES-16 or E is just a single snap shop that reveals the cloud structures over a very dynamic North Atlantic. We can combine that with low earth orbiting data from the METOP-A satellite scatterometer and can have an good indication of the surface wind field.

Displays Easily interpreted Data acquisition time QC or appropriate flagging Can distinguish criteria (Beaufort scale) Standardized displays Observations Numerical model output Can thin for legibility Be careful not to thin relevent signals We make decisions based on a single piece of data To optimize displays we worked with our forecasters to create displays that are easily interpreted. Specific wind speed ranges are colored coded with anything yellow and warmer are critical criteria requiring wind warnings. Ranges match the Beaufort categories with yellow being GALE and and orange Strong GALE. Brown is storm with reddish brown being violent storm Force 11.

We copied those displays for numerical model output so that our forecasters could easily overlay satterometer winds over the model analysis or short term forecast and do a quick assessment of model short term skill. In this example the winds field of storm force is larger than the US GFS and stronger with the scatterometer winds of violent storm force.

Hurricane Force Extra-tropical Cyclones - Detection and Warning Trend using QuikSCAT 2000-2009 Hurricane Force Warning Initiated Dec 2000 Detection increased with: Forecaster familiarity Data availability Improved resolution Improved algorithm Improved wind algorithm and rain flag Oct 06 12.5 km QuikSCAT available May 04 25 km QuikSCAT Available in N-AWIPS Oct 01 Totals A-289 P-269 558 Hurricane Force Wind Warning Initiated Dec 00 QuikSCAT Launch Jun 99

Extreme Maritime Weather Storms of Hurricane Strength ATL PAC 2017-18 47 26 2018-19 39 30 Location of storms that produced Hurricane force winds The number of extreme storms is phenomenal We are able to observe these conditions via the Metop scatterometers via consistent and standardized ocean vector winds measurements that have been validated using buoys and NOAA P-3 flight data. 6

GOES-16 Sea Surface Temperature We looked at clouds and winds and this example is meant to illustrate the sea surface temperature and wind relationship. First – the 2 km IR capability of the GOES-R series with the ABI imager is a big advancement in being able to being to observe the complexity and dynamic nature of the ocean surface. This is a 4 day composite of SST with the most recent pixel being displayed… the starting point is one hourly image with the volume being 24 per day over 4 days. We are changing that to 10 min imagery over 4 day period and displaying the most recent to take full advantage of the capability. I do think that this capability is underappreciated and we will grow with it as it become available and cloud separation schemes improve and become more definitive. SST is a challenge to display as fixed ranges over the year can mask ocean features… more dynamic capabilities to highlight features such as SST fronts is really needed but is also dependent on use.

GOES-16 Sea Surface Temperature In this example the Wind speeds from ASCAT-B show a blotchy wind field with large blobs of higher winds from the Gulf Stream / North Atlantic Drift southward over the warmer water. The the north of the GS the winds are much lower. This is due to well mixed PBL over the warmer water and a highly stratified PBL with a very shallow inversion over the cooler water restricting mixing.

Applications Wind Warnings Oil Detection Phytoplankton Sea Ice Hurricane Force Winds Wind Warnings Oil Detection Phytoplankton Sea Ice Sea Ice Coverage Coral Reef Bleaching Limit Bycatch Snow and Sea Ice Parameters Ocean Vector Winds Waves Sea Surface Height Sea Ice SAR NRCS Sea Surface Temperature Ocean Color/Chlorophyll Eddy Kinetic Energy Sampling of applications of imagery and the parameters and combinations of parameters observed/retrieved. Mammal Protection Harmful Algal Blooms

Training Hands on Integrated displays data Interaction w/SMEs Students models observations Interaction w/SMEs Students depart with data and application No guarantee post course Training efforts sponsored by EUMETSAT and supported by NOAA via contributing instructors. This class was last Dec in Centurion South Africa at the temporary SAWS facility. A good facility with segregated server for the training room to most efficiently access and display data and use a common display software package.

Thoughts… What are spatial and temporal sampling needs for phenomena of interest? Orbit phasing, repeat cycle, diurnal cycle Having similar sensors Low data latency Data sharing of satellite capabilities to enhance or create a constellation Altimeters Ocean winds Usage – forecasters make critical decision based on one data point requires display of representative resolution (instrument capability) Validation Example – ocean winds in extreme conditions – P-3 flights Integrated observations (collocation in space and time) of atmos and ocean Determine dependencies (EX: Wind, SST, Rain) Training Co-located with Subject Matter Experts (not so for most) Requires validated and consistent data (similar answer from similar instruments) Data access and readability needed – understand strength and weaknesses Post training – access and utilization (continues a big challenge)